A Survey on Sentiment Analysis and Opinion Mining: A need for an Organization and Requirement of a customer

Sentiment analysis and Opinion mining is the computational study of User opinion to analyze the social, psychological, philosophical, behavior and perception of an individual person or a group of people about a product, policy, services and specific situations using Machine learning technique. Machine learning for text analysis technically has always been very challenging as its main goal is to make computers able to learn and automatically generate emotions like a human as it is practically very useful in real life scenarios. After a boom in web 2.0 technology this field became the most interesting area for researchers because the social media has grown as the fastest medium for availability of opinions. There are many commercial tools available in the market and many researchers have proposed their solutions for opinion extraction, but still there are some problems of text classification and sentiment extraction in opinion mining. These problems arise due to different behaviors, manners and textual habits of users. A sentence can be positive for one ,but it may have a negative impact on other so it will be a problem for a machine to generate its emotion. A negative sentence can be written in a positive manner like “What a great camera! It consumes more battery power, this sentence has a negative opinion about a camera, but it consists only positive keywords. There are mainly four predominating problems viz. subjectivity classification, word sentiment classification, document sentiment classification and opinion extraction. Data mining algorithms are easy to implement, but concludes to poor accuracy meanwhile the machine learning technique provides better accuracy, but requires a lot of training time, so there should be a hybrid technique which has the advantages of both the techniques. This survey focuses on various Data Mining algorithms, Machine learning techniques and a brief review stating the comparative analysis of these algorithms. We have followed a systematic literature review process to conduct this survey and also mentioned the future aspects of sentiment analysis and opinion mining. Keyword – Sentiment Analysis, Opinion Analysis, Market Research, machine Learning